Global AI Newsletter·Issue 22

On April 24, the People's Bank of China, the Ministry of Industry and Information Technology, the State Administration for Market Regulation, the National Financial Regulatory Administration, the China Securities Regulatory Commission, CNIPA, the Cyberspace Administration of China, and the State Administration of Foreign Exchange jointly issued the Measures for Online Marketing of Financial Products (hereinafter the "Measures"), effective from September 30, 2026. The Measures stipulate that when using algorithm recommendation technology for online marketing, entities must not establish algorithm models that induce financial consumers and investors to overconsume, and must simultaneously provide options that do not target individual characteristics, or provide convenient options to disable algorithm recommendation services.

Link:https://www.gov.cn/zhengce/zhengceku/202604/content_7066927.htm

2. Ministry of Commerce and SAMR Issue Action Plan for Services Trade Standardization (2026-2030)

On April 23, the Ministry of Commerce website published the Action Plan for Services Trade Standardization (2026-2030)(hereinafter the "Action Plan"),with the overall goal of basically establishing by 2030 a standards system covering multiple fields and the full chain of services trade, adapted to high-quality services trade development needs, expanding standards supply, enhancing international influence, and strengthening China's participation in international services trade standardization governance.

The Action Plan sets specific requirements for digital trade standards, prioritizing standards development and system construction in areas such as software and information services, data security and privacy protection in digitally delivered trade, cross-border data element flows, and digital trade platform construction, exploring standards for service scenarios deeply integrated with AI, virtual reality, and other new technologies, conducting standardization practices for AI-empowered services trade, and accelerating the improvement of the institutional and market environment for digital trade development.

Link:https://fms.mofcom.gov.cn/zcfg/zhzcfg/art/2026/art_7ccf1e930cda41aab74fa8f7963f4742.html

3. MIIT Proposes Accelerating Construction of About 20 Specialized Large Models and High-Quality Datasets in Consumer Goods

On April 21, the State Council Information Office held a press conference on Q1 industrial and information development. MIIT spokesperson Tao Qing stated that the ministry will deepen the "AI + Three Quality" special action, planning to build about 20 consumer goods specialized large models covering home furnishing, apparel, food, cosmetics, home appliances and other sectors, while simultaneously building high-quality industry datasets to address underlying technical shortcomings.

Meanwhile, MIIT will develop application guides for AI empowerment in key industries, guiding enterprises to develop flexible production, intelligent supply chains, personalized customization and other new models, creating immersive new consumption scenarios, promoting AI from the R&D end to the application end, effectively enhancing consumer experience, supporting quality improvement and efficiency enhancement and transformation and upgrading of the consumer goods industry, and providing implementation support for deep integration of the digital and real economies.

Link:https://www.digitalchina.gov.cn/2026/xwzx/szkx/202604/t20260423_5313382.htm

4. National Archives Administration Launches AI Application Pilot Program

On April 20,the General Office of the National Archives Administration issued the《Notice on Conducting AI Application Pilot Work,》deciding to organize AI application pilot work to steadily and orderly advance AI technology applications in the archives industry. The pilot aims to implement the national "AI+" action, promote digital transformation and intelligent upgrading of archives work, explore replicable and scalable experiences across six dimensions: computing power, algorithms, data, scenarios, security, and standards.

The pilot tasks primarily include five aspects: first, exploring the construction of regional computing centers or self-built computing platforms and establishing computing resource scheduling management mechanisms; second, building general archives large models and strengthening model selection and localized deployment; third, creating high-quality archives datasets and improving data annotation, quality inspection, and training/testing sets; fourth, deepening existing application scenarios and creating new business scenarios to drive digital-intelligent transformation of archives operations; fifth, constructing security management and control systems covering data compliance, model training, algorithm vulnerabilities, and ethical security.

Link:https://www.saac.gov.cn/daj/tzgg/202604/ca1a627cfee844fe9abbbb147afc2c36.shtml

5. CNIPA Strengthens IP Protection for AI and Other Emerging Industries

On April 24, the State Council Information Office held a press conference introducing the Annual Report on China's Efforts to Combat Infringement and Counterfeiting (2025). CNIPA spokesperson and Director of the IP Protection Department Guo Wen stated that CNIPA has taken a series of measures to strengthen the development of AI and other strategic emerging industries: first, establishing dedicated chapters in the Patent Examination Guidelines to standardize examination criteria for AI and big data, strengthening AI ethics review, and clarifying requirements for AI-related data collection and rule-setting; second, optimizing patent examination policies by providing multiple channels including priority examination and expedited examination, dynamically updating product and service items, with 890 items related to big data, AI and other new industries and business models now publicly disclosed; third, establishing 82 national-level IP protection centers nationwide, covering frontier directions and key sectors including AI, integrated circuits, quantum technology, and brain-computer interfaces; fourth, building 116 overseas dispute response guidance platforms and strengthening early warning monitoring and risk screening.

Link: http://www.scio.gov.cn/live/2026/38437/index.html?fromModule=lemma_middle-info

6. Shanghai Issues Data Export Negative List Management Measures and Negative List for Pilot Free Trade Zone and National Services Opening-Up Comprehensive Pilot Area

On April 24, the Shanghai Municipal Internet Information Office and Shanghai Municipal Data Bureau, together with relevant departments, formulated the Management Measures (Trial) and Data Export Management List (Negative List) (2025 Version) for the China (Shanghai) Pilot Free Trade Zone, Lingang Special Area, and National Services Opening-Up Comprehensive Pilot Area. The Negative List covers 4 industry sectors including reinsurance, international shipping, commerce and trade, and meteorology, involving 9 specific scenarios, 29 data subcategories, and 109 data items. Compared to the free trade zone negative list released on February 8, 2025, this list refines the specific data items for the international shipping negative list, serves the needs of local meteorological model overseas deployment, and adds a new meteorology sector negative list.

Link: https://mp.weixin.qq.com/s/h1fKq2XE6CgqK-eCTjgWvA

7.Guangdong Provincial Government Office Issues Action Plan for Accelerating AI Application Across All Domains, All Hours, and All Industries

On April 22, the Guangdong Provincial Government Office issued the Action Plan for Accelerating High-Level AI Application Across All Domains, All Hours, and All Industries（(hereinafter the "Plan").）。

《The Plan》consists of three parts: overall requirements, key directions, and factor support. Key directions include“"AI +"”scientific research,“"AI +"”agriculture and rural areas,“"AI +"”traditional industries,“"AI +"”emerging industries,“"AI +"”services,“"AI +"”governance capabilities, and“"AI +"”people's well-being across seven areas. Factor supportincludes the followingpriorities:First,strengthening computing power coordination;；Second,enhancing data supply;；Third,fostering open-source ecosystem;；Fourth,building talent teams;；Fifth,securing the safety foundation.

Link:https://www.cnbayarea.org.cn/policy/policyrelease/policies/content/post_1324890.html

8.Shanghai Municipal Government Office Publicly Issues Implementation Plan for National Digital Economy Innovation Development Pilot Zone (Shanghai)

On April 21, the Shanghai Municipal Government Office publicly issued the Implementation Plan for the National Digital Economy Innovation Development Pilot Zone (Shanghai), proposing targets including full activation of data factor value and further deepening of digital-real economy integration by 2028, deploying thirty tasks across seven areas.

Key measures include: exploring data property rights systems and public data authorized operation pricing mechanisms, creating high-quality datasets focusing on healthcare, high-end manufacturing and other sectors; advancing National Blockchain Network (Shanghai Hub) construction, upgrading smart computing infrastructure, accelerating Qianfan Constellation and low-altitude intelligent connected systems; pursuing breakthroughs in ultra-heterogeneous fusion chips, homomorphic encryption, 6G, quantum computing and other frontier technologies; promoting digital transformation in shipping and trade, blockchain-empowered cross-border RMB clearing and settlement; supporting Pudong, Xuhui and other districts in building national data industry clusters, cultivating 50 benchmark digital economy enterprises; expanding data export negative list applicability to all of Shanghai, building Lingang International Data Economy Industrial Park at high standards, developing“"inbound data processing"”and other new international data trade models.

Link:https://www.shanghai.gov.cn/nw12344/20260421/9edf5d5c1b2244298c315304349d0963.html

9. Shanghai Issues Action Plan for Industrial Internet Platform Empowering Industrial Development (2026-2028)

On April 20, Shanghai issued the Action Plan for Promoting Industrial Internet Platform Empowerment of Industrial Development (2026-2028)（(hereinafter the "Action Plan").）The Action Plan includes four main tasks: first, strengthening platform industrial empowerment leadership; second, building stronger platform digital-intelligent technology capabilities; third, constructing platform internationalization service systems; fourth, creating a quality ecosystem for platform development.

Link: https://www.sheitc.sh.gov.cn/cyfz/20260420/ac9220e475984b5bb37efb51ecfae29d.html

10. Yangzhou Jiangdu District at“"AI +"”Innovation Development Conference Unveils Three Digital-Intelligent Economy Centers

On April 25,“the 2026 Jiangdu "AI+" Innovation Development Conference”was held in Jiangdu, Yangzhou. At the conference, the Jiangdu Digital-Intelligent Economy Out-of-Area (Hangzhou) Industry Promotion Center, the Yangzhou Jiangdu District AI Development Center, and the Data Factor Comprehensive Service Center were officially unveiled. The Out-of-Area Industry Promotion Center aims to connect with resource spillover from digitally advanced regions such as Hangzhou; the AI Development Center focuses on local enterprise intelligent transformation and digital upgrading; the Data Factor Service Center is dedicated to promoting integrated applications of public and enterprise data. The conference also featured multiple industry application roadshows and centralized signing of 14 digital-intelligent economy projects covering intelligent connected vehicles, industrial internet, and other sectors.

Link: https://www.js.chinanews.com.cn/news/2026/0425/233330.html

1.2 Law Enforcement and Judicial Updates

1. SPC Issues Judicial Interpretation on Punitive Damages in IP Infringement Civil Disputes

4April20, the Supreme People's Courtissuedthe Interpretation on Applying Punitive Damages in Civil Dispute Cases Involving IP Infringement (Judicial Interpretation [2026] No. 7).The Interpretation further refines legal application standards for key issues in applying punitive damages in IP judicial practice.

First, it further refines the criteria for determining "intent" and "serious circumstances," adding situations such as "after reaching a settlement with the plaintiff and agreeing to cease infringement, engaging in the same or similar infringing conduct again" as circumstances from which the defendant's intent to infringe IP can be determined, further clarifying the meaning of "engaging in IP infringement as a business," and refining rules for determining serious IP infringement conduct in accordance with law.

Second, it further clarifies the calculation method for the base amount. It specifies that when using the defendant's illegal gains or infringement profits as the punitive damages base, operating profit may be referenced; for defendants engaging in IP infringement as a business, sales profit may be referenced; when profit margins cannot be determined, the average profit margin or the right holder's profit margin for the same period and industry published by statistics departments or industry associations may be referenced. It clarifies that statutory damages cannot serve as the base for punitive damages calculation.

Third, it improves the method for determining the multiplier. Based on the principle of proportionality, it specifies that when the same infringing conduct has already been subject to administrative fines or criminal penalties that have been executed, the court shall take this into consideration when determining the punitive damages multiplier, without requiring the parties to submit a request.

Link: https://www.court.gov.cn/zixun/xiangqing/497911.html

2. SPC Issues People's Courts IP Judicial Protection Implementation Plan (2026-2030)

4April20, the Supreme People's Court issued the People's Courts IP Judicial Protection Implementation Plan (2026-2030)(hereinafter the "Implementation Plan"). The Implementation Plan includes twenty-two specific measures covering strengthening protection of scientific and technological innovation to serve high-level scientific and technological self-reliance, strengthening copyright protection to promote socialist cultural prosperity, strengthening judicial protection of data rights to serve high-quality digital economy development, and properly adjudicating AI-related cases to promote beneficial, safe, and fair development.

Link: https://www.court.gov.cn/zixun/xiangqing/497921.html

3. SPC Issues Summary of Key Rulings of the SPC IP Court (2025)

On April 22, the Supreme People's Court issued the Summary of Key Rulings of the SPC IP Court (2025) (hereinafter the "Summary"). The Summary selected 132 cases from 3,146 cases concluded by the SPC IP Court in 2025, extracting 159 key rulings covering eight judicial practice areas: patent authorization and validity, patent ownership and infringement, plant variety rights, integrated circuit layout design, trade secrets, computer software, antitrust, and procedural cases, collectively showcasing the SPC IP Court's judicial philosophy, adjudication approaches, and ruling methods in technology-related IP and antitrust cases for reference by all sectors of society. With few exceptions, the relevant judicial documents have been published on the China Judgments Online website.

Link:https://www.court.gov.cn/zixun/xiangqing/498181.html

4. SPC Issues Report on IP Judicial Protection in Chinese Courts (2025)

4April20, the Supreme People's Court held a press conference to release the Report on IP Judicial Protection in Chinese Courts (2025)(hereinafter the "Report"). The Report comprehensively presents the achievements of IP adjudication quality and efficiency, judicial policy supply, key area protection, digital economy and AI governance, civil-criminal-administrative coordination, and international exchange and cooperation across all courts nationwide in 2025, organized around six major themes: serving scientific and technological innovation and new quality productive forces, maintaining fair competition, strengthening copyright and trademark protection, advancing diversified dispute resolution, deepening adjudication reform, and enhancing foreign-related judicial capabilities.

Link:https://www.court.gov.cn/zixun/xiangqing/497931.html

5. SPC Publishes Ten Typical IP Cases of 2025

On April 20, the Supreme People's Court held the 2026 IP Awareness Week press conference and published ten typical IP cases of 2025. The typical cases cover ten key areas: distinctiveness determination of phrase-type trademarks, chip invention patent infringement, refurbished switch resale trademark infringement, improper acquisition of trade secrets through internal-external collusion, unfair competition through malicious talent poaching, unfair competition through scraping network platform data, counterfeiting registered trademarks, commercial defamation, copyright infringement through online sale of pirated e-books, and unfair competition through repeated malicious trademark registration.

Link: https://www.court.gov.cn/zixun/xiangqing/497941.html

6. First National AI Open Source Ecosystem Consensus Released in Guangzhou

On April 20, the Guangdong High Court held a symposium themed "Judicial Escort for Innovation - Open Source Co-Governance for Win-Win," where 24 organizations in the national AI field jointly released the Consensus on Strengthening Collaborative Innovation to Promote the Prosperity of the AI Open Source Ecosystem. The consensus clearly recognizes the legal validity of open source licenses, commits to standardizing use, modification, derivative development, and commercialization within the licensing framework, and calls on the industry to jointly resist plagiarism, tampering, and theft of achievements that damage the ecosystem. It particularly emphasizes protecting the legitimate rights of foundation large model open source parties, respecting their investments in computing power, data, R&D, and maintenance, and promoting more quality technologies to embrace open source.

Link:https://www.news.cn/local/20260421/4592d0d51bb6488c91342691b5a6fa23/c.html

7. Beijing IP Court Publishes Typical AI-Related Cases

On April 21, the Beijing IP Court held its 2025 annual case release conference, publishing 10 annual cases and 9 nominated cases. Among them, Case 5 involves an unfair competition dispute over "transformation manga special effects," marking the first case involving protection of AI model and parameters. The case clarifies that the AI model structure and parameters formed through data training, optimization, and calibration, which can bring innovation advantages and operating income to the operator, constitute legitimate interests protected by the Anti-Unfair Competition Law, and unauthorized commercial use of such model structures and parameters by others constitutes unfair competition.

Link:https://bjzcfy.bjcourt.gov.cn/article/detail/2026/04/id/9283798.shtml

8. Beijing High Court Publishes Typical AI-Related Cases

On April 22, the Beijing High Court held a press conference on "IP Judicial Services Supporting New Quality Productive Forces Development," publishing typical cases of Beijing courts serving and supporting new quality productive forces development in 2025. Case 5 involves a virtual digital human copyright infringement dispute, clarifying that a virtual digital human may be recognized as a fine art work if it reflects the creative team's unique aesthetic choices and judgments. Case 7 involves a criminal case of using AI generative models to infringe copyright, clarifying that using AI as a tool to reproduce and distribute others' original fine art works without authorization for profit constitutes the crime of copyright infringement.

Link:https://mp.weixin.qq.com/s/PSey48bRJPSzGTwOBRGmYQ

9. Shanghai High Court Publishes Typical AI-Related Cases

On April 23, the Shanghai High Court held a press conference publishing typical IP and trade secret protection cases of Shanghai courts in 2025. Case 7 involves an AI search engine infringing on the right of dissemination through information networks, clarifying the dynamic requirements for "technology neutrality" infringement exemption for search engine service providers developed based on large language model retrieval-augmented generation under the premise of AI algorithm transparency. Case 8 involves an AI prompt copyright infringement dispute, where the prompts in question, while reflecting certain creative intent, did not demonstrate the author's personalized intellectual input at the expression level and were not recognized as works.

Link:https://mp.weixin.qq.com/s/_fMXjdSEYf1m3r1-6fulFg

10. Beijing-Tianjin-Hebei Three High Courts Jointly Publish Typical AI-Related Cases

On April 26, the copyright bureaus of Beijing, Tianjin, and Hebei jointly hosted the 2026 National IP Awareness Week Beijing (Beijing-Tianjin-Hebei) Copyright Theme Event, with the three high courts jointly publishing "Typical Cases of Digital Copyright Judicial Protection in Beijing-Tianjin-Hebei Courts." Case 1 involves a pendant image "AI text-to-image" copyright infringement dispute,，where the court clarified that the determination of originality of AI-generated outputs should require the creator to submit prompts, iteration processes, sketches, selection records, modification records, and other evidence to support their rights claim. Case 3 involves an "AI face-swapping" copyright infringement dispute, where the court regulated the use of AI face-swapping technology to partially synthesize others' works, thereby infringing on the copyright of original short video works.

Link: https://www.hebeicourt.gov.cn/article/detail/2026/04/id/9289505.shtml

11. Hangzhou Intermediate Court Publishes Typical AI-Related Cases

On April 22, the Hangzhou press conference on building a pilot city for optimizing the IP business environment was held, where the Hangzhou Intermediate Court published the "2025 Annual Typical Cases of IP Judicial Protection 'Three-in-One' in Hangzhou Courts." Case 3 involves a dispute between Sun and Company A over infringement of the right of dissemination through information networks. The judgment, without denying the value of generative AI as a creative tool, clarifies the standard for determining the originality of generative AI-generated content by analyzing the impact of prompts on originality determination in the "text-to-image" function of generative AI.

Link:https://mp.weixin.qq.com/s/nXlmgGLShY1Eqa6WY07_DQ

12. Beijing Disposes of Batch of Illegal Financial Accounts

4April21, the Beijing Municipal Internet Information Office, together with the Beijing Financial Regulatory Bureau, steadily advanced the "Qinglang Jinghua - Financial Guardian" special campaign, directing local websites and platforms to clean up over 155,000 pieces of illegal information and dispose of over 39,000 illegal accounts in accordance with law, including four typical cases involving impersonating professionals to mislead the public, publishing illegal referral information in covert ways, promoting and referring illegal deposit and lending intermediary services, and disseminating illegal agency and other illegal information.

Link: https://mp.weixin.qq.com/s/kTcCShjRyPtU4q3fJqXQEg

1.3 International Cooperation Updates

1.China and Mozambique Sign MoU on Strengthening Digital Economy Cooperation and Two Other Cooperation Documents

On April 21, witnessed by the heads of state of China and Mozambique, the National Development and Reform Commission and relevant Mozambican departments signed three cooperation documents on jointly building the“Belt and Road,”economic development cooperation, and digital economy, deepening practical cooperation between China and Mozambique in development strategy alignment and the digital domain. Specifically: NDRC Chairman Zheng Shanjie and Mozambican Minister of Planning and Development Vala signed the Departmental MoU on Key Cooperation Items for Jointly Building the“Belt and Road”and the MoU on Exchange and Cooperation in Economic Development; National Data Administration Director Liu Liehong and Mozambican Minister of Communications and Digital Transformation Muchanga signed the Departmental MoU on Strengthening Digital Economy Cooperation.

Link:https://www.nda.gov.cn/sjj/jgsz/jld/llh/llhldhd/0422/20260422222733496250122_pc.html

1.4 Research Updates

1. World Internet Conference Asia-Pacific Summit Releases Five AI Governance Reports

On April 22,，the World Internet Conference official websitereported thatduring the Asia-Pacific Summit's“Smart Benefits for People's Livelihoods Forum,”five research reports focusing on AI governance were releasedcollectively,including the report on Global Intelligent Divide: Evolutionary Trends, Multi-Dimensional Impacts, and Cooperative Governance Pathways released by Zhejiang Lab's Intelligent Social Governance Lab; the report on Global Intelligent Divide: Evolutionary Trends and Inclusive Pathways jointly released by Zhejiang University International Communication Research Center and Wuzhen Digital Civilization Research Institute; the report on Network Information Content Governance in the Generative AI Era released by Renmin University of China Interdisciplinary Research Institute; the report on Global AI Governance Legislative Observations: Experiences and Outlook jointly released by Wuhan University Competition Law and Policy Research Center, Xinjiang University Law School, and Wuhan University Network Governance Research Institute; and the Digital Intelligence Empowering Government Governance Evaluation Index report released by Nankai University Network Social Governance Research Center.The content covers global intelligent divide, network information content governance in the generative AI era, global AI governance legislative observations, and digital intelligence empowering government governance evaluation.。

Link:https://cn.wicinternet.org/2026-04/22/content_38725178.htm

1.5 Industry Updates

1. DeepSeek Open-Sources and Launches V4 Model

On April 24, DeepSeek announced the launch and open-sourcing of its new-generation foundation model DeepSeek-V4.

DeepSeek-V4 comes in two versions by size: Pro and Flash. The Pro version has total parameters of 1.6T with 49B activated parameters, making it the largest open-weight model to date; the Flash version has total parameters of 284B with 13B activated parameters. Both models feature million-token context capacity. According to media reports, DeepSeek-V4 pioneered a novel attention mechanism that compresses along the token dimension, combined with DeepSeek Sparse Attention (DSA), achieving world-leading long-context capabilities while significantly reducing computing and memory requirements compared to traditional methods. From now on, 1M (one million) context will be standard across all DeepSeek official services.

Link:https://www.ithome.com/0/942/955.htm

2. 2026 China Database Technology and Industry Conference Held in Beijing

On April 22, the 2026 China Database Technology and Industry Conference was held in Beijing, themed "Dameng Building Foundations, Intelligence Opening New Era." China Software Industry Association Vice Chairman Yang Ju noted that China's database industry has transitioned from following to running alongside and leading, with domestic databases achieving core business support in finance, telecommunications, and other key industries, and the industry shifting toward a "technology + product + ecosystem" three-dimensional leadership pattern.

At the conference, Dameng Data released four new products: Database Management System DM9, new-generation database appliance DAMENG PAI V2.0, Dameng Qiyun Database V4.0, and Graph Database GDMBASE V4.0, covering centralized, distributed, and cloud-native core scenarios. The conference also launched a Domestic Database Industry-University-Research Innovation Consortium to support collaborative industrial innovation.

Link:https://www.chinanews.com/cj/2026/04-22/10608891.shtml

3. China（(Nanjing)）Software Valley Signs Strategic Cooperation Framework Agreement with CAICT，Focusing on Digital Employee Platform Construction

On April23, China（(Nanjing)）Software Valley and the China Academy of Information and Communications Technology formally signed a strategic cooperation framework agreement. Both parties will closely align with the national "AI+" action deployment and the“"AI +"”policy orientation of“"building new forms of intelligent economy,"”focusing on digital employee platform construction and OPC（"one-person company"）model innovation, building a new cooperation system of "platform-based support, scenario-based implementation, industrialized incubation, and governance-based operation"; planning to co-build a standardized digital employee supply system, integrating underlying technologies into a reusable and scalable intelligent service matrix, establishing OPC scenario adaptation standards covering the full lifecycle, and solving compliance and stability challenges of one-person company intelligent transformation.

Link:https://www.news.cn/tech/20260423/4fc0fed6754c44349869d294b64c64b6/c.html

4. Shanghai Forum Sub-Forum "Global South Digital Economy Connectivity: Silk Road E-Commerce and Diverse Practices" Held

On April 25, the Shanghai Forum sub-forum "Global South Digital Economy Connectivity: 'Silk Road E-Commerce' and Diverse Practices" was held at Fudan University. The forum conducted in-depth discussions around three major themes: first, national digital strategy and policy dialogue, analyzing Silk Road countries' digital agendas and "Silk Road E-Commerce" international cooperation pathways; second, technology innovation and industrial application, focusing on next-generation intelligent agent transformation, cross-border e-commerce comprehensive pilot zone advantages, and new models for cross-border data flows, with participating experts noting that "Silk Road E-Commerce" has transcended the Silk Road countries' scope to build a broader cooperation ecosystem; third, regional practices and localization experience, sharing digital opportunities and challenges in Indonesia, ASEAN, and the Balkans.

Link:https://finance.sina.com.cn/jjxw/2026-04-25/doc-inhvtryx0524597.shtml

5. 2026 Asia-Pacific Digital Economy Forum Seminar Held in Beijing

On April 25, the 2026 Asia-Pacific Digital Economy Forum“"Northeast Asia - Free Trade Port Digital Intelligence Corridor Initiative"”Seminar was held in Beijing, where participants launched an initiative to jointly promote“"data for good, bridging the data divide, and unlocking data value."”The organizing committee conducted systematic discussions on how to leverage Jilin Changchun's rich data resource endowment, Beijing's digital intelligence technology talent advantages, and Hainan Free Trade Port's high-level data cross-border opening policies to create a new form of digital-intelligent economy in the Northeast Asia-—ASEAN region. During the seminar, the Asia-Pacific Digital Economy Forum organizing committee, together with the Hong Kong Polytechnic University Digital Economy Research Center, Xi'an Jiaotong University Data Factor Research Center, Jilin University School of AI, and participating data enterprises, jointly launched the Northeast Asia - Free Trade Port Digital Intelligence Corridor Initiative.

Link:https://www.chinanews.com.cn/cj/2026/04-25/10610778.shtml

II. International Governance Updates

2.1 Policy and Legislative Updates

1. Netherlands Publishes AI Act Implementation Draft Establishing Decentralized Regulatory Model

On April 20, the Dutch government published the draft Implementation Act on the AI Act (Uitvoeringswet AI-verordening), defining the national AI regulatory architecture. The Netherlands did not establish a unified AI regulatory body but adopted a "decentralized supervision" model, with 8 existing sector regulators responsible for AI compliance oversight in their respective areas such as finance, healthcare, and infrastructure. Meanwhile, the Dutch Data Protection Authority (AP) and the State Inspectorate for Digital Infrastructure (RDI) serve coordinating functions, with RDI as the single external contact point.

The draft emphasizes preventing regulatory fragmentation through mandatory cooperation mechanisms, requiring regulatory bodies to establish information sharing and coordinated enforcement mechanisms and unified interpretation of key AI Act concepts. In terms of enforcement, regulators have powers including investigation, remediation, suspension or prohibition of AI system use, and imposition of substantial fines.

Link: https://www.internetconsultatie.nl/uaiv/b1

2. Two U.S. House Committees Introduce SECURE Data Act and GUARD Financial Data Act

On April 22, the U.S. House Committee on Energy and Commerce and the House Committee on Financial Services introduced the SECURE Data Act and the GUARD Financial Data Act respectively. According to committee statements, the former primarily establishes nationwide privacy and data security rules covering general personal data, while the latter focuses on personal information protection in financial contexts, updating financial data protections under the Gramm-Leach-Bliley Act (GLBA) framework.

From a regulatory design perspective, the two bills primarily reflect four aspects: first, strengthening consumer data rights - the SECURE Data Act would grant consumers rights to know, access, delete, and opt out of targeted advertising, personal data sales, and certain automated decisions, requiring user consent for sensitive data processing and parental consent for children and adolescents; second, strengthening enterprise data minimization and security obligations - requiring enterprises to limit personal data collection to what is "adequate, relevant, and reasonably necessary for disclosed purposes" with corresponding data security measures; third, enhancing data broker regulation - data brokers must register with the FTC and disclose their privacy and data security practices and types of personal data sold, with the FTC to establish a publicly searchable data broker registry; fourth, emphasizing scenario-specific financial data protection through the GUARD Financial Data Act, updating requirements for access, use, and disclosure of data held by financial institutions.

Link:https://energycommerce.house.gov/posts/committees-on-energy-and-commerce-and-financial-services-introduce-pair-of-privacy-bills-to-establish-comprehensive-data-protections-for-all-americans

3. Australia's eSafety Commissioner and Office of the Australian Information Commissioner Sign MoU on Privacy and Cybersecurity

On April 23, the Office of the eSafety Commissioner (eSafety) and the Office of the Australian Information Commissioner (OAIC) signed a memorandum of understanding to strengthen cooperation in protecting Australian citizens' online privacy and cybersecurity.

Under the MoU, the two regulators will establish more formal coordination mechanisms building on existing cooperation to address the growing intersection of privacy and cybersecurity issues. Key focuses include: first, establishing formal communication and coordination mechanisms to ensure aligned action on regulatory matters involving privacy and cybersecurity; second, conducting joint work on mandatory age verification requirements to protect children from online abuse and inappropriate content; third, jointly overseeing social media platforms' compliance with minimum age obligations; fourth, promoting industry deployment of age verification technologies that meet regulatory requirements while respecting user privacy in new technology (including AI) application scenarios.

Link: https://www.esafety.gov.au/newsroom/media-releases/esafety-and-oaic-working-together-to-protect-privacy-and-safety-for-all-australians

4. Singapore Develops International Standard Draft for Generative AI Testing

On April 20, Singapore's Infocomm Media Development Authority (，IMDA) announced that Singaporewillrelease a new international standard draft at the 17thISO/IEC JTC 1/SC 42Plenary Assembly held from April 20 to 24 (SO/IEC AWI TS 42119-8). This is the first international standard on generative AI testing, aiming to provide standardized testing services for generative AI. The standardfocuses on benchmarking and red-teaming methodologies for generative AI systems, employing standardized testing methods.The standard proposes definitions, requirements, and guidance for generative AI testing,covering quality assessments (including safety assessments) conducted using red-teaming and other methods.。Overall, thestandardestablishes an important AI testing framework that enhances the reproducibility and comparability of results.

Link:https://www.imda.gov.sg/resources/press-releases-factsheets-and-speeches/press-releases/2026/singapore-champions-new-global-ai-testing-standardisation-efforts

5. European Commission Allocates EUR 63.2 Million for Health and Online Safety AI Innovation

On April 21, the European Commission announced that under the Digital Europe Programme (Digital Europe Programme), it will launch 7 calls for proposals worth a total of EUR 63.2 million,aimed at supporting AI in healthcare, digital health, digital skills, and cybersecurity.Key projects include: first, EUR 9 million for developing AI image screening technologies in healthcare centers, an area with great potential for improving prevention, early detection, and treatment of cancer and cardiovascular diseases; second, EUR 24 million to supportdigital healthcare services and systems under the European Health Data Space;third, EUR12.5 millionfor funding advanced digital skills training across the EU, supporting skills development needed for Europe's modern economy; fourth, EUR8.5 millionfor innovative digital solutions to promote EU regulatory compliance.

Link:https://digital-strategy.ec.europa.eu/en/news/commission-makes-eu632-million-available-support-ai-innovation-health-and-online-safety

6. Ghana Releases National AI Strategy to Drive Digital Economy Transformation

On April 24, Ghanaian President Mahama officially released the National AI Strategy in Accra.（Ghana’s National Artificial Intelligence Strategy）。The strategy proposes thatGhana aims to become an AI hub for West Africa and the continent, focusing on priority industries including healthcare, agriculture, education, finance, energy, security, and public services, following the principle of "people-centered, augmentation not replacement." To advance strategy implementation, Ghana will invest $250 million to build a national AI computing center, with an additional $20 million for short-to-medium-term implementation. The government has established AI liaisons in each ministry, introduced performance assessment indicators, and is cultivating digital talent through the "Million Programmers Initiative." The strategy is built on eight pillars including AI education, digital infrastructure, data governance, and industry applications, led by the Ministry of Communications, Digital Technology and Innovation.

Link:https://moc.gov.gh/2026/04/24/ghana-launches-national-ai-strategy-to-drive-digital-transformation-and-economic-growth/#

7. Vietnamese PM Deploys Strategic Core Technology Priority Investment and Sets Digital Economy Targets

On April 22, Vietnamese Prime Minister Le Minh Khai emphasized at a science and technology work conference the priority allocation of resources to，increase the share of strategic technologies in national fiscal expenditure. According to the CPV Central Committee resolution, science, technology, innovation, and digital transformation should receive no less than 3% of national budget expenditure annually,，andtargets are set for the digital economy to reach 30% of GDP and for the proportion of enterprises participating in innovation to exceed 40%.

Le Minh Khai deployed specific tasks: requiring the Ministry of Science and Technology to submit 3 decrees and 2 prime ministerial decisions within April to guide the implementation of the Digital Transformation Law and AI Law; improving evaluation and funding mechanisms for S&T projects, promoting outcome-based commissioned funding, establishing controlled R&D regulatory sandboxes; drafting a new national development model resolution based on S&T and innovation; advancing reform of 16 national key laboratories, strengthening IP valuation, and completing innovation center and innovative enterprise certification standards by May 2026.

Link:https://en.vietnamplus.vn/pm-underscores-priority-funding-for-strategic-core-technologies-post341598.vnp

8. Canada's Privacy Commissioner Comments on Connected Care for Canadians Act

On April 22, the Office of the Privacy Commissioner of Canada,（Office of the Privacy Commissioner of Canada，OPC）Deputy Commissioner Marc Cheniere, made a presentation to the Senate committee on the Connected Care for Canadians Act (Bill S-5). The OPC supports a more interoperable digital healthcare system but emphasizes that health information, as one of the most sensitive types of personal information, must place privacy and security at the core of related systems.

Specifically, the OPC's comments focus on three areas: first, affirming the bill's direction on interoperability while emphasizing that access, use, and exchange of personal health information should not cross existing privacy law boundaries, particularly noting that the bill's interoperability provisions already specify that access, use, and exchange of personal health information is not required where prohibited by privacy law, and recommending that the proposed "data blocking" prohibition rules should include similar exceptions to prevent digital connectivity obligations from inversely compressing privacy protection space; second, noting that the bill leaves substantial specific requirements and standards for subsequent regulation, including privacy and security requirements, and that privacy regulatory authorities should be fully consulted in subsequent rulemaking; third, emphasizing continued coordination with provincial and territorial privacy regulators to promote rule alignment, and reiterating the importance of modernizing the Personal Information Protection and Electronic Documents Act (PIPEDA) to ensure more adequate protection for the private sector including health technology.

Link:https://www.priv.gc.ca/en/opc-actions-and-decisions/advice-to-parliament/2026/parl_260422/

9. UK Security Minister Calls on AI Companies to Co-Build National Cyber Defense Capabilities

On April 22, UK Security Minister Dan Jarvis delivered a keynote speech at the CYBERUK cybersecurity conference in Glasgow, calling on leading AI companies to work with the government to jointly develop AI-driven national cyber defense capabilities.

UK National Cyber Security Centre (NCSC) data shows that in 2025, the number of significant national security incidents handled by the center more than doubled compared to previous years, with hostile states and criminals increasingly using automated AI systems to identify and exploit vulnerabilities. In response, Jarvis simultaneously launched the voluntary "Cyber Resilience Pledge," inviting companies to raise security standards through three specific actions: first, incorporating cybersecurity into board-level responsibility and implementing the Cyber Governance Code of Practice; second, registering for NCSC's free early warning service; third, requiring supply chain partners to obtain government-backed Cyber Essentials certification. The pledge, developed by the Department for Science, Innovation and Technology (DSIT), will be formally launched this summer.

Meanwhile, the UK government announced an additional GBP 90 million investment over three years, focusing on strengthening cyber resilience for small and medium enterprises (SMEs). This funding will be implemented through existing DSIT and NCSC programs. These measures will be incorporated into the National Cyber Action Plan to be released this summer.

Link: https://www.gov.uk/government/news/call-to-action-for-ai-companies-to-work-with-uk-government-on-national-cyber-defence

2.2 Law Enforcement and Judicial Updates

1. Italian Privacy Regulator Imposes Heavy Fines on Postal Financial Institutions for Data Violations

On April 20, the Italian Data Protection Authority issued enforcement decisions, fining Poste Italiane S.p.A. EUR 6.624 million and Postepay S.p.A. EUR 5.877 million, totaling over EUR 12.5 million, for illegally processing millions of users' personal data in their mobile applications.

The investigation revealed that both apps made "monitoring user phone data (including installed and running applications)" a mandatory condition for using services, used to identify malware and prevent fraud. However, the regulator determined this practice violated the principles of necessity and proportionality, constituting excessive data collection.

Additionally, the regulator found multiple compliance issues including: inadequate user notification, failure to conduct data protection impact assessments, insufficient security measures, missing data retention policies, and irregular data processing relationships.

In addition to fines, the regulator ordered both companies to cease the relevant data processing and undertake comprehensive remediation.

Link:https://www.garanteprivacy.it/home/docweb/-/docweb-display/docweb/10241568?

2. Alabama Attorney General Reaches Settlement with Roblox to Strengthen Minor Protection

On April 21, the Alabama Attorney General announced a settlement agreement with Roblox totaling $12.2 million to address issues related to minor protection on its platform. According to the Alabama AG's office, the settlement not only requires Roblox to make a series of improvements to platform governance, but Alabama will also receive the full settlement funds to support school resource officers statewide through the AG's Safe Schools Initiative. The AG's office also stated that the agreement establishes an enforceable minor protection framework for platforms that can serve as a reference for other states.

The settlement includes systematic constraints around platform product design, content management, parental controls, and minor data protection: first, regarding communication safety, Roblox must require all users to complete age confirmation beyond self-reported age before using chat functions by May 1, limit chat between adult users and users under 16 unless they are "trusted friends," issue risk alerts for first private messages to minors, continuously improve communication safety measures, and submit annual safety improvement reports to the Alabama AG's office; second, regarding content safety, Roblox must set default content modes for users under 16 and unverified users, restricting access to only screened age-appropriate content, opening "restricted" or 18+ content only to age-verified adults, requiring all discoverable game content to have clear maturity ratings understandable to parents, with measures including warnings, removal, permission restrictions, and bans for developers who falsely label content; third, regarding parental controls, Roblox must provide remote management tools and visual dashboards through linked parent accounts covering communication, contacts, content restrictions, spending limits, privacy settings, and screen time, with personalized ads and mobile push notifications disabled by default for users under 16, and nighttime notification restrictions for users aged 13-15; fourth, regarding personal information protection, the agreement requires encryption, hashing, least-privilege access, multi-factor authentication, and other protective measures for information collected during age verification, requires third-party vendors to delete data collected for age verification as soon as reasonably practicable, and prohibits use of such data for purposes beyond age determination, fraud prevention, and user safety.

Link:https://www.alabamaag.gov/wp-content/uploads/2026/04/2026-04-21_-_Roblox_AL_-_Final_Term_Sheet_-_for_Roblox_signature.pdf

3. Ontario Information and Privacy Commissioner Publishes PowerSchool Case Key Findings

On April 20, the Information and Privacy Commissioner of Ontario (IPC) published a "lessons learned" case study on the PowerSchool data breach, systematically reviewing its investigation findings concerning 20 Ontario school district education boards and the Ontario Ministry of Education regarding a data breach at an education technology platform. According to the IPC's disclosure, PowerSchool Canada ULC, as an education technology service provider, provides data management services for students and educators from kindergarten through grade 12; in January 2025, relevant public institutions reported to the IPC that their education technology vendor had suffered a cyberattack resulting in the breach of students' and educators' personal information.

The IPC primarily identified four areas of concern: first, public institutions' contract governance and ongoing oversight of third-party education technology providers was clearly insufficient, with some contract terms not being robust enough and institutions failing to establish regular, formal oversight mechanisms to verify PowerSchool's ongoing compliance with privacy and security obligations or to adequately require annual risk assessments, audits, or compliance reports; second, technical security controls had clear shortcomings, with attackers exploiting compromised high-privilege accounts, and insufficient log retention and lack of mandatory multi-factor authentication for support portals expanding risk exposure; third, over-collection and over-retention issues, with some schools collecting personal information not necessary for educational purposes and some breached data traceable back decades, indicating that minimum necessary collection and reasonable retention period controls were not effectively implemented; fourth, outsourcing does not exempt public institutions from data protection responsibilities. The IPC ultimately issued 14 recommendations and required relevant institutions to submit compliance or remediation progress reports within 6 months.

Link:https://www.ipc.on.ca/en/cases-of-note/millions-student-records-compromised-why-overseeing-education-technology-companies-crucial?utm_source.com

4. Oregon Appellate Court Sanctions Lawyer for Submitting AI-Generated False Content

According to Reuterson April 23report,the Oregon Court of Appeals ruled that lawyers using generative AI for legal work should proactively disclose to judges when errors result. The court sanctioned the plaintiff's attorney in a defamation case for submitting a complaint containing AI-fabricated quotes and other errors. The plaintiff's attorney admitted using generative AI as a research and drafting aid but submitted some content without independent verification. The court ordered the plaintiff's attorney to pay approximately $8,000 in attorney's fees to the defendant to compensate for costs incurred in responding to the defective filing. This was the first time the appellate court sanctioned AI-caused false legal materials through fee-shifting rather than fines.

Link: https://www.reuters.com/legal/government/lawyers-should-disclose-when-ai-causes-errors-appeals-court-says-2026-04-23/

2.3 International Cooperation Updates

1. ILO Develops First Global Framework for AI Application in Manufacturing

On April 21, the International Labour Organization (ILO) released the first-ever global "tripartite conclusions" on AI in manufacturing. Adopted on April 17 following discussions by government, employer, and worker representatives from 54 countries, the conclusions aim to guide stakeholders in harnessing AI potential while limiting negative impacts in the manufacturing sector employing nearly 500 million workers.

The conclusions cover multiple dimensions of decent work: respecting and implementing fundamental principles and rights at work, employment, skills and occupational safety and health policies, social protection, and the role of social dialogue in managing AI-driven change. Key recommendations include: strengthening skills development and lifelong learning; creating an enabling environment for enterprises to use AI to enhance productivity; ensuring fundamental rights including occupational safety and health; safeguarding decent working conditions; improving AI-related regulatory frameworks; and strengthening social dialogue to manage transitions.

Link:https://www.ilo.org/resource/news/ilo-adopts-first-ever-conclusions-ai-manufacturing-work

2. India-KoreaLaunch "Digital Bridge" Project to Strengthen AI and Chip Cooperation

On April 20, Indian Prime Minister Modi and South Korean President Lee Jae-myung released the Joint Strategic Vision for India-ROK Special Strategic Partnership (2026-2030) in New Delhi. The core outcome is the launch of the“India-Korea Digital Bridge”framework, focusing on AI, data governance, and digital business, specifying semiconductorsand otherenabling technologies as support for digital innovation. Both sides will deepen AI cooperation across all areas, including joint research and talent development, and advance faculty and student exchange programs in AI and STEM at universities.。

On industrial cooperation, a new Industrial Cooperation Committee is established, focusing on semiconductor, automotive, shipbuilding, telecommunications equipment, secondary battery, and other supply chain coordination, and strengthening critical mineral supply chain resilience. The two sides also announced acceleration of the Comprehensive Economic Partnership Agreement upgrade negotiations and promotion of digital payment system interconnectivity. In defense, the“Korea-India Defense Accelerator”innovation platform is launched, while in space, cooperation between India's ISRO and Korea's KASA is promoted.

Link: https://www.pib.gov.in/PressReleasePage.aspx?PRID=2253977&ref=finshots.in&reg=3&lang=2

3. Commonwealth Launches Regional Training on AI and Electoral Integrity in Trinidad and Tobago

On April 20, the Commonwealth launched a regional training program in Trinidad and Tobago for the Caribbean and Americas region, themed "Artificial Intelligence: Opportunities & Risks for Electoral Integrity in the Commonwealth Caribbean & Americas."。

This five-day training is part of the Commonwealth Electoral Professionals（Commonwealth Election Professionals，CEP）initiative, focusing on AI's dual impact on elections: exploring the use of AI to improve electoral efficiency while guarding against risks such as misinformation and algorithmic bias that undermine electoral integrity. Discussion topics include ethical issues in AI-driven elections, legal regulatory considerations, and risk management. The training aims to help Electoral Management Bodies (EMBs) build proactive response capabilities and develop electoral integrity protection strategies for the digital age,，ensuring transparent, trusted, and inclusive electoral processes.

Link:https://thecommonwealth.org/news/commonwealth-launches-regional-training-artificial-intelligence-and-electoral-integrity

4. 2026 Digital World Conference Held in Geneva

According to Xinhua on April 24, the 2026 Digital World Conference was held on April 21 at the Palais des Nations in Geneva, Switzerland. The conference, themed "AI Driving Social Development," brought together government officials, academics, and international organization representatives from around the world, focusing on topics including AI governance, social security, green transition, education and training, and culture and arts.。Participating experts unanimously agreed that AI has significant potential for driving economic and social development but also brings new challenges to governance systems, social equity, and public trust, emphasizing the need to strengthen global coordinated governance to ensure AI development is oriented toward social development and inclusivity. Shen Jian, Deputy Representative of China's Permanent Mission in Geneva, noted in his remarks that the importance of AI in global development is increasingly prominent, calling on the international community to strengthen cooperation in infrastructure construction, capacity building, and technology transfer, with particular attention to the development needs of developing countries and marginalized groups.

Link:https://www.news.cn/world/20260422/51fcb76d8a5a408aa26a2ad9a398ead1/c.html

5. IDB Group Launches New Caribbean Cybersecurity Regional Cooperation Initiative at ONE Caribbean Dialogue

On April 21, the Inter-American Development Bank Group (IDB Group) announced three new cooperative actions around regional shared challenges during the "ONE Caribbean" ministerial dialogue, the most directly relevant to data security and digital governance being the cybersecurity preparedness and rapid response mechanism. According to the IDB Group website, this round of new actions focuses on three areas: cybersecurity, capital market integration, and fiscal research.

The IDB Group explicitly stated that under the Cybersecurity Preparedness and Rapid Response initiative being advanced in cooperation with the CARICOM Implementation Agency for Crime and Security (IMPACS), it is providing targeted support to member states to develop cybersecurity action plans and has established a dedicated rapid technical assistance facility for responding to serious cybersecurity incidents.

Link: https://www.iadb.org/en/news/idb-group-advances-regional-cooperation-one-caribbean-ministerial-dialogue

2.4 Research Updates

1. WAIFC Releases 2026 Annual Report on AI in Financial Services

On April 21, the World Alliance of International Financial Centres (WAIFC), in cooperation with Abu Dhabi Global Market (ADGM), released the 2026Annual AIReport,（WAIFC AI Report 2026）based on a comprehensive survey conducted across 12 member jurisdictions, exploring five key themes: the impact of AI on financial services, regulatory considerations, talent development, ecosystem innovation, and related risks and opportunities.The report notesthat AI applications have become firmly embedded in International Financial Centres (IFCs), widely applied in compliance, fraud detection, customer service, and portfolio management. While most jurisdictions have data protection laws applicable to AI, only a few have introduced AI-specific regulations. Many IFCs are aligning with international standards, but gaps remain in areas such as accountability for AI-driven violations and autonomous decision-making.

Link:https://assets.adgm.com/download/assets/WAIFC+AI+Report+2025.pdf/bd126c9a3e4311f1bbeb1a145cc763dd

2.5 Industry Updates

1. Zimbabwe Hosts 2026 Digital Economy Conference

On April 21, the Ministry of Information Communication Technology, Postal and Courier Services (MICTPCS) of Zimbabwe hosted the "2026 Digital Economy Conference" in Bulawayo, themed "Accelerating Zimbabwe's Digital Economy: Harnessing Artificial Intelligence for Action Towards Vision 2030." The conference aimed to bring together global thought leaders, industry experts, and innovative enterprise representatives for exchanges on topics including the evolving digital economy landscape, AI applications, and digital transformation, through keynote speeches, panel discussions, roundtable dialogues, hands-on training, and networking, promoting collaboration among policymakers, industry, and academia to provide digital momentum for Zimbabwe's Vision 2030 national development goals.

Link:https://digitaleconomyconference.org.zw/

ꄴPrevious： null

ꄲNext： null